I’m using TVRX2 and uhd_fft to capture the spectrum of a 6 mhz wide QAM
signal:
uhd_fft -s 10M -f 603M --args addr=192.168.10.3 --spec A:RX1 -g 1
–averaging
The QAM carrier is centered on 603 mhz. If you look at the attached
snapshot
of the uhd_fft output, note the null right in the middle of the QAM
carrier, at
603 mhz. Any idea where this is coming from? I tried the obvious
tweaks
of adjusting the input level, padding the input to TVRX2, to no avail.
Both inputs to the TXRV2 seem to suffer the same issue.
Note, there are adjacent QAM carriers at +/- 6 mhz from the desired
carrier
at 603 mhz.
Input level in the 603 mhz channel is probably about -10 dbmV.
Perhaps a related question: the gain uhd_fft supplies to the block seems
overly sensitive. TVRX2 uses a tuner designed for CATV service, so
given a normal input level of around 0 dbmV in an application like CATV,
I would have expected a more midrange gain setting. Going much above
a gain of 1-3 seems to result in compression of the tuner, i.e. the
observed
CNR (just judging by eyeballing the uhd_fft output) seems to reduce
dramatically.
Any ideas about where else to look on the null problem? Tx Rick